An alternative strategy to the Pictet-Spengler method for tetrahydroisoquinoline synthesis: A feasibility study

Article abstract of DOI:10.1016/j.tetlet.2012.06.091

Acid-catalysed cyclisations of the 2-aminoethyl styrene derivatives 9 give good to excellent yields of the corresponding tetrahydroisoquinolines 7 by an intramolecular hydroamination reaction, which represents an alternative and potentially more flexible strategy to the classical Pictet-Spengler method for the syntheses of such heterocycles.

Full text of DOI:10.1016/j.tetlet.2012.06.091

Tetrahedron Letters 53 (2012) 4657–4660
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
An alternative strategy to the Pictet–Spengler method
for tetrahydroisoquinoline synthesis: a feasibility study
Laura Henderson a, David W. Knight a, , Andrew C. Williams b
⇑
a School of Chemistry, Cardiff University, Main College, Park Place, Cardiff CF10 3AT, UK
b Eli Lilly & Co. Ltd, Lilly Research Centre, Erl Wood Manor, Windlesham, Surrey GU20 6PH, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 26 June 2012
Acid-catalysed cyclisations of the 2-aminoethyl styrene derivatives 9 give good to excellent yields of the
corresponding tetrahydroisoquinolines 7 by an intramolecular hydroamination reaction, which repre-
sents an alternative and potentially more flexible strategy to the classical Pictet–Spengler method for
the syntheses of such heterocycles.
Keywords:
Pictet–Spengler
Alternative
Ó 2012 Elsevier Ltd. All rights reserved.
Hydroamination
Acid-catalysed
Cyclisation
The Pictet–Spengler reaction1 belongs to that venerable group
of classical heterocyclic synthetic methods which have served their
purpose extremely well over many years, despite some quite sig-
nificant limitations. A generalised mechanism consists of an initial
condensation between a phenethylamine 1 and an aldehyde, fol-
lowed by activation of the resulting imine using a Brønsted or Le-
wis acid, and cyclisation of the resulting activated electrophilic
species 2, to arrive at the tetrahydroisoquinoline product 3
(Scheme 1).2 Naturally, more recently there has been considerable
interest and success in desymmetrizing the central cyclisation step
to produce chiral, non-racemic products 3, along with the related
products derived from indole precursors, especially tryptophans.3
It has been known for over 30 years that such chemistry is also
used by Nature in the biosynthesis of many tetrahydroisoquinoline
residues; in a notable development, the enzymes responsible for
this transformation, the aptly named Pictet–Spenglerases, have
been isolated and their structures determined.4,5 There is, however,
a significant limitation to the Pictet–Spengler reaction, in that it re-
quires an activated, electron-rich aryl ring to complete the cyclisa-
tion. This has certainly not restricted its many applications in
alkaloid synthesis, precisely because a very high proportion of
these natural product targets possess exactly such an activated aryl
ring, being either methoxylated or hydroxylated para to the de-
sired reaction site. Even with such activating groups present, the
formation of lesser amounts of the ortho-substituted product 4
can sometimes be an additional irritation. Thus, while this neces-
sary condition usually does not affect the viability of the reaction
MeO
MeO
MeO
RCHO
H+
NH2
N
H
R
1
2
3
H
R
N
MeO
NH
R
4
Scheme 1. The classical Pictet–Spengler reaction.
TfOH
R1
R2
R1
R2
CH2Cl2
N
TsHN
Ts
5
6
Scheme 2. An acid-catalysed pyrrolidine synthesis.
in such target synthesis, it can severely restrict its applications in
the preparation of non-natural relatives. In view of the exception-
ally important biological activities displayed by many of these
alkaloids, this is obviously an area of great potential for the
⇑
Corresponding author.
E-mail address: knightdw@cf.ac.uk (D.W. Knight).
0040-4039/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetlet.2012.06.091

4658
L. Henderson et al. / Tetrahedron Letters 53 (2012) 4657–4660
discovery of novel pharmaceuticals which, in general, cannot
NHTs
NHTs
therefore be accessed using the Pictet–Spengler method. Herein,
we wish to report an alternative strategy which has the potential
to overcome this limitation. While still based around a reactive
and electron-deficient intermediate, this new approach to
tetrahydroisoquinolines already has the two aryl substituents in
place and hence cannot suffer from the foregoing limitations.
The idea arose from our discovery that alkenyl sulfonamides 5
could be induced to cyclise upon exposure to strong acid,
NTs
R
R
R
7
8
9
Scheme 3. Our proposed alternative to the Pictet–Spengler reaction.
Table 1
Acid-catalysed cyclisations of styrenes 9 to give N-sulfonyl-tetrahydroisoquinolines 7a
Entry
Precursor
Conditions
Product
Yield (%)
NHTs
NHTs
NTs
0.5 equiv TfOH, 0 °C, 24 h
0.4 equiv H2SO4, 20 °C, 48 h
100
99
1
Ph
Ph
NTs
2
0.5 equiv TfOH, 40 °C, 24 h
87
Cl
Cl
NHTs
NTs
3
0.5 equiv TfOH, 40 °C, 24 h
72
F
F
NHTs
NTs
0.5 equiv TfOH, 40 °C, 48 h
0.4 equiv 4, 40 °C, 48 h
75
56
4
F3C
CF3
NHTs
NTs
0.5 equiv TfOH, 40 °C, 24 h
0.4 equiv H2SO4, 40 °C, 48 h
54
80
5
6
NHTs
NTs
0.5 equiv TfOH, 40 °C, 24 h
90
NHTs
NHTs
NTs
NTs
0.5 equiv TfOH, 0 °C, 24 h
0.4 equiv H2SO4, 20 °C, 24 h
83
81
7
8
0.5 equiv TfOH, 0 °C, 24 h
0.4 equiv H2SO4, 40 °C, 24 h
92
89
a
All reactions were carried out in stirred, anhydrous dichloromethane.

L. Henderson et al. / Tetrahedron Letters 53 (2012) 4657–4660
4659
specifically catalytic trifluoromethanesulfonic (triflic) acid or con-
centrated sulfuric acid.6 Although clearly a novel pyrrolidine
synthesis, this type of transformation belongs to the burgeoning
area of hydroamination methodologies, and should be especially
suited to the synthesis of highly substituted and sterically crowded
amine derivatives.7
The resulting pyrrolidines 6 were generally isolated in excep-
tionally high yields, but usually, where relevant, as mixtures of
stereoisomers. (Scheme 2). We have more recently applied this
found in the conditions necessary to drive the cyclisations to com-
pletion in the case of these aryl substituted (stilbene) examples. In
general, concentrated sulfuric acid was less reactive than triflic
acid, not surprisingly, but this still gave quite similar yields of
the cyclised products. The alkenyl substituted precursors (entries
6–8) reacted somewhat more rapidly but equally cleanly, with
the exception of the cyclohexenyl-substituted precursor in entry
7, which understandably reacted more rapidly, presumably be-
cause the reactive intermediate is a tertiary benzylic carbenium
ion. The lack of very rapid cyclisation is presumably a reflection
of the considerable degree of steric hindrance present in this exam-
ple, which is of considerable significance as it demonstrates that
this methodology can be used to obtain such spiro-derivatives
which would be difficult to prepare using alternative methods. It
should be noted that the reaction times reported in Table 1 have
not been optimised.
methodology to a total synthesis of the pentacyclic alkaloid
a-
cyclopiazonic acid, in which a key transformation was a cascade
cyclisation onto an initial benzylic carbenium ion, generated by
exposure of a secondary benzylic alcohol to acid.8 Having thus
established that benzylic carbenium ions were able to participate
in such cyclisations, we wondered if somewhat related chemistry
could be successfully applied to the synthesis of tetrahydroiso-
quinolines; the basic idea is outlined in Scheme 3. Thus, if one
imagines heterolytic cleavage of the C–N bond in a tetrahydroiso-
quinoline 7, then the resulting carbenium ion intermediate 8 might
be generated from the corresponding styrene 9 upon exposure to
catalytic acid. Of course, protonation of the alkene function in pre-
cursors 9 could also lead to the isomeric carbenium ion, especially
when R = Ar and thence to formation of a seven-membered ring.
This possibility, along with determination of the exact conditions
required to efficiently achieve the desired alkene protonation were
our initial concerns, the successful resolution of which we report
herein.
We therefore suggest that this methodology should form the
basis of a viable alternative to the classical Pictet–Spengler reac-
tion. Further studies aimed at more fully defining its scope, stereo-
chemical features and applicability to other (hetero)aromatic
frameworks are underway.
Acknowledgement
We are grateful to Eli Lilly and Co Ltd and the EPSRC for finan-
cial support.
Very recently, a similar disconnection, but using carbamates
rather than sulfonamides as the nucleophiles and carbenium ions
generated by acidification of benzylic alcohols has been shown to
be very useful for the synthesis of tetrahydroisoquinolines. How-
ever, all examples reported contained a 3,4-dimethoxyphenyl
group, in which the presumed carbenium ion was in conjugation
with one of the methoxy substituents and thus could suffer from
exactly the same limitation as the original Pictet–Spengler reaction
(see above).9 1-Vinyl-tetrahydroisoquinolines have also been ob-
tained by formation of the same C–N bond by overall SN20 mecha-
nisms, again using nucleophilic attack by a carbamate group but in
References and notes
1. Pictet, A.; Spengler, T. Ber. Dtsch. Chem. Ges. 1911, 44, 2030–2036.
2. Whaley, W. M.; Govindachari, T. R. Org. React. 1951, 6, 151–190; Rozwadowska,
M. D. Heterocycles 1994, 39, 903–931; Cox, E. D.; Cook, J. M. Chem. Rev. 1995, 95,
1797–1842; Scott, J. D.; Williams, R. M. Chem. Rev. 2002, 102, 1669–1730;
Chrzanowska, M.; Rozwadowska, M. D. Chem. Rev. 2004, 104, 3341–3370;
Larghi, E. L.; Amongera, M.; Bracca, A. B. J.; Kaufman, T. S. ARKIVOC 2005, 98–
153.
3. Lorenz, M.; Van Linn, M. L.; Cook, J. M. Curr. Org. Synth. 2010, 7, 189–233. and
references therein.
4. For an excellent review of this aspect of the Pictet–Spengler reaction, as well as
recent developments in asymmetric synthesis, see Stöckigt, J.; Antonchick, A.
P.; Wu, F.; Waldmann, H. Angew. Chem., Int. Ed. 2011, 50, 8538–8564.
5. For some very recent applications and developments in both natural product
synthesis and materials chemistry, see Dong, W.; Liu, W.; Liao, X.; Guan, B.;
Chen, S.; Liu, Z. J. Org. Chem. 2011, 76, 5363–5368; Yokoya, M.; Shinada-Fujino,
K.; Saito, N. Tetrahedron Lett. 2011, 52, 2446–2449; Mastranzo, V. M.; Yuste, F.;
Ortiz, B.; Sanchez-Obregon, R.; Toscano, R. A.; Garcia Ruano, J. L. J. Org. Chem.
2011, 76, 5036–5041; Shumaila, A. M. A.; Puranik, V. G.; Kusurkar, R. S.
Tetrahedron Lett. 2011, 52, 2661–2663.
6. Griffiths-Jones, C. M.; Knight, D. W. Tetrahedron 2010, 66, 4150–4166;
Schlummer, B.; Hartwig, J. F. Org. Lett. 2002, 4, 1471–1474; For the
development of an asymmetric version using a chiral, non-racemic yttrium
catalyst, see Chapurina, Y.; Ibrahim, H.; Guillot, R.; Kolodziej, E.; Collin, J.;
Trifonov, A.; Schulz, E.; Hannedouche, J. J. Org. Chem. 2011, 76, 10163–10172.
7. For recent reviews of alkene hydroamination, see Müller, T. E.; Beller, M. Chem.
Rev. 1998, 98, 675–703; Nobis, M.; Driessen-Hölscher, B. Angew. Chem., Int. Ed.
2001, 40, 3983–3985; Pohlki, F.; Doye, S. Chem. Soc. Rev. 2003, 32, 104–114;
Müller, T. E.; Hultzsch, K. C.; Yus, M.; Foubelo, F.; Tada, M. Chem. Rev. 2008, 108,
3795–3892; Majumdar, K. C.; Debnath, P.; De, N.; Roy, B. Curr. Org. Chem. 2011,
15, 1760–1801; For alkyne hydroaminations, see Severin, R.; Doye, S. Chem. Soc.
Rev. 2007, 36, 1407–1420; Patil, N. T.; Singh, V. J. Organomet. Chem. 2011, 696,
419–432.
these cases onto a p
-allylic palladium intermediate10 or a bismuth-
stabilized allylic carbenium ion.11 Further, the former method
delivers chiral, non-racemic products when a chiral phosphine is
attached to the metal. In view of these reports, we wish to present
our own preliminary results in this area, which feature largely suc-
cessful outcomes to the idea shown in Scheme 3.
As described in the foregoing paper,12 the necessary precursors
9 are readily accessible from optimised Suzuki–Miyaura couplings
between vinylboronic acids and N-tosyl-2-bromophenethylamine,
using microwave activation, in company with a pre-mixed catalyst
system. We were delighted to find that exposure of the precursors
9 to sub-stoichiometric quantities of either triflic acid or concen-
trated sulfuric acid in dichloromethane resulted in slow but
smooth cyclisation, as anticipated in Scheme 3, to give generally
good to outstanding yields of the desired tetrahydroisoquinolines.
The results are collected in Table 1.13
8. Griffiths-Jones, C. M.; Knight, D. W. Tetrahedron 2011, 67, 8515–8528.
9. Ivanov, I.; Nikolova, S.; Aladjov, D.; Stefanova, I.; Zagorchev, P. Molecules 2011,
16, 7019–7042.
10. Chien, C.-W.; Shi, C.; Lin, C.-F.; Ojima, I. Tetrahedron 2011, 67, 6513–6523.
11. Kawai, N.; Abe, R.; Matsuda, M.; Uenishi, J. J. Org. Chem. 2011, 76, 2102–2114.
12. Henderson, L.; Knight, D. W.; Williams, A. C. Tetrahedron Lett. 2012, 53,
preceding paper.
In general, the cyclisations were slow but clean when using
around half an equivalent of the acid catalyst.13 In the first five
examples, there is a distinct possibility of formation of the isomeric
benzylic carbenium ion, trapping of which by the sulfonamide
would lead to the corresponding seven-membered azepine prod-
ucts. This is presumably especially true in the case of the relatively
electron-rich aryl substituents present in entries 1 and 5. However,
in none of these examples were any traces of such seven mem-
bered products identified by 1H NMR analysis of the crude prod-
ucts. Although not fully optimised, there was little difference
13.
A
typical procedure is as follows:- 1-benzyl-2-(4-toluenesulfonyl)-1,2,3,4,-
tetrahydroisoquinoline (entry 1, Table 1): (E)-4-methyl-N-[2-(2-
phenylethenyl)phenethyl]benzenesulfonamide (97 mg, 0.257 mmol) was
dissolved in dry dichloromethane (10 ml) to which was added concentrated
sulfuric acid (ca. 35 mg, two drops). The resulting heterogeneous mixture was
stirred at ambient temperature for 48 h then basified using a slight excess of
2 M aqueous potassium carbonate. The separated aqueous layer was extracted

4660
L. Henderson et al. / Tetrahedron Letters 53 (2012) 4657–4660
with dichloromethane (2 ꢀ 10 ml) and the combined organic solutions dried
2.34 (m, 1H), 2.23 (s, 3H, TsMe); dC (CDCl3, 100 MHz) 143.0 (C), 137.7 (C), 136.9
(C), 135.6 (C), 133.5 (C), 129.9 (2 ꢀ CH), 129.5 (2 ꢀ CH), 128.7 (CH), 128.3
(2 ꢀ CH and CH), 127.3 (CH), 127.2 (2 ꢀ CH), 126.9 (CH), 125.0 (CH), 57.9
over potassium carbonate and filtered through a pad of silica gel. Evaporation
of the filtrate and washings left the isoquinoline (96 mg, 99%) as a pale yellow
oil which showed dH (CDCl3, 400 MHz) 7.37 (d, 2H, J 8.3 Hz), 7.13–6.85 (m,
10H), 6.75–6.71 (m, 1H), 5.12 (app. t, 1H, J 6.6 Hz, CHN), 3.48–3.42 (m, 1H,
CHaHb), 3.36–3.28 (m, 1H, CHaHb), 3.07–3.02 (m, 2H), 2.64–2.55 (m, 1H), 2.42–
(CHN), 44.5 (CH2N), 40.0 (CH2), 27.2 (CH2), 21.5 (TsMe); m
max/cmꢁ1 (film) 3290,
3060, 3027, 1652, 1598, 1448, 1274, 990, 815; HRMS (ES) m/z calcd for
C
23H24NO2S, [M+H]+, 378.1528; found 378.1539.